同步操作将从 科学大数据开源社区/图数据库系统-gStore 强制同步,此操作会覆盖自 Fork 仓库以来所做的任何修改,且无法恢复!!!
确定后同步将在后台操作,完成时将刷新页面,请耐心等待。
This chapter introduces some useful tricks if you are using gStore to implement applications.
If you are using gStore to serve as a SPARQL Endpoint, you had better set the SPARQL_ENDPOINT macro in Util.h. In addition, all DEBUG macros in Util.h should not be used. What is more, if you do not use gStore as endpoint, but you do not need to update the database, ONLY_READ macro should be set in Util.h.
When running as a HTTP server, gStore has provided a backup function, and you can modify the time interval of backup procedure in Util. However, this backup utility can not take effect if the whole disk is down. As a result, you had better use multiple machines or clouds to do other backup procedures by yourself, if you are requiring a high security.
When running as a HTTP server, gStore has set a time limit for query processing, i.e. 1 hour. You can modify this parameter in Util as you wish, but we suggest that the lowest bound is 1 minute.
The efficiency of KVstore has huge impact on the performance of the whole system, and you can modify related parameters in KVstore.h, according to your demand and memory capacity.
To speed up the process of reading Strings from disk when running into getFinalResult() function of the query processing, gStore has provided String Buffers for entities and literals. You can set this parameters in setStringBuffer() of Database.h according to the memory capacity of your machine.
If you are using HTTP API, and use Java to visit it, then you must be care for the efficiency(due to the Garbage Collection process in jvm). We strongly suggest that you read codes in api/http/java carefully, before you write your own program to use the REST API providing by HTTP protocol.
此处可能存在不合适展示的内容,页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容,可点击提交进行申诉,我们将尽快为您处理。